#asahi-gpu · 2025-04-10 17:57
jannau`vkmark -bvertex` has regressed slightly: ~5.9k (old uapi) vs. 5.4k (new uapi). Testing a different install though #asahi-gpu · 2025-04-10 17:26
jannauglmark2 raw value of 6.13 is lower than 6.14 old uapi but probably in the margin of error#asahi-gpu · 2025-04-10 17:22
jannauI'll run vkmark#asahi-gpu · 2025-04-10 17:22
jannau6.13 seems to be roughly the same as 6.14 with old uapi#asahi-gpu · 2025-04-10 17:20
jannaum1 max#asahi-gpu · 2025-04-10 17:16
jannaulest me try 6.13.8#asahi-gpu · 2025-04-10 17:15
jannaueven the 4.1k are lower than my recollection#asahi-gpu · 2025-04-10 17:15
jannauold vs. new uapi on asahi-6.14.1-2 has glmak score of ~4.1k vs. 2.7k#asahi-gpu · 2025-04-09 21:19
jannau\o/#asahi-gpu · 2025-03-14 21:18
jannauas it takes control away from the driver. on the other hand the only slternative the driver has is to let the coprocessor crash#asahi-gpu · 2025-03-14 21:17
jannauI'm not sure if that's a good solution but handling it once in the iommu driver seemed good trade-off vs implementing it in multiple device drivers. not convinced it's a good solution#asahi-gpu · 2025-03-14 21:12
jannauthe drivers do not necessarily need to handle the reserved memory but we just need to maintain the mappings apple's bootloader sets up when it loads the firmware#asahi-gpu · 2025-03-14 21:10
jannaurobher: I'm currently cheating with of_iommu_get_resv_regions() with changes in https://github.com/AsahiLinux/linux/commit/01db68e6333139a16c3010cfeee9d8a8d45e6df7 / https://github.com/AsahiLinux/linux/commit/704ace01cd3c423b1e2492f0777d9c4c0f3404d8 #asahi-gpu · 2025-02-27 18:30
jannaurobher: page tables and a handoff region. the agx coprocessor like all coprocessors gets initialized by iboot#asahi-gpu · 2025-02-25 23:10
jannauI still have to replace the xarray abstractions#asahi-gpu · 2025-02-25 23:09
jannauthis is after abstactions already under discussion on rust-for-linux#asahi-gpu · 2025-02-25 23:07
jannauless than 2.5k added rust abstractions left (excluding drm): https://paste.centos.org/view/325a8533#asahi-gpu · 2025-02-25 22:22
jannauand working with adding more hacks#asahi-gpu · 2025-02-25 21:50
jannauI suppose this should be used by all other asc drivers as well instead of the the automatic mapping I implemented for dart#asahi-gpu · 2025-02-25 21:47
jannauin principle this should have all timed used of_reservedmem but that looks annoying to wrap for rust#asahi-gpu · 2025-02-25 21:44
jannaudoesn't work yet because reserved CPU memory shouldn't be handled in the same way as iomem#asahi-gpu · 2025-02-25 20:30
jannauagx driver compiles again after replacing a ton stale rust dependencies with upstream changes or patches under active discussion#asahi-gpu · 2024-08-22 18:50
jannauwhen will the default SSD be to small to hold all ail tests?#asahi-gpu · 2024-07-27 12:40
jannauthe offical position is that if you can't figure that out on your own what you need to do it's not ready yet#asahi-gpu · 2024-07-27 12:38
jannauplease do not make up non-existing / not yet existing tags#asahi-gpu · 2024-07-27 12:37
jannaumostly unrelated to the uapi changes we will push in the next few days#asahi-gpu · 2024-07-27 12:37
jannauhoneykrisp is merged in upstream mesa, so it still needs at least need a rebase of asahi/mesa but I would imagine that it will not build by default for some time after that#asahi-gpu · 2024-04-26 12:19
jannauI've asked in kwin's matrix channel#asahi-gpu · 2024-04-26 11:47
jannaumy bet would be a plasma problem on asahi#asahi-gpu · 2024-04-26 11:36
jannauchadmed_: oh, I assumed this might be an issue with my config and didn't have a chance to investigate. the panels fix themself when you switch to a different virtual desktop#asahi-gpu · 2023-09-16 08:00
jannauif easy to determine `rebase --onto asahi/main asahi/$prev_main` should work better. asahi/$prev_main could be a tag or commit id#asahi-gpu · 2023-09-16 07:59
jannaui509vcb: that probably tries to rebase commits from the previous asahi/main onto the current one#asahi-gpu · 2023-09-13 04:37
jannauah, I can't count (or don't care for iPhone SoCs). M class SoCs don't appear to be on a yearly schedule so I would expect them to skip an A* generation from time to time#asahi-gpu · 2023-09-12 18:32
jannauthat's just PR for all we know#asahi-gpu · 2023-09-12 18:31
jannaualso "the biggest GPU redesign in Apple's history"#asahi-gpu · 2023-09-12 18:06
jannaumesh shading and ray tracing in g15 (pro?)#asahi-gpu · 2023-09-08 17:37
jannauit would allow using multi page folios for mmap but it's far off from being merged and might not be application controllable#asahi-gpu · 2023-09-08 17:33
jannaudone#asahi-gpu · 2023-09-08 17:33
jannauI can send an subscriber link#asahi-gpu · 2023-09-08 17:32
jannauhttps://lwn.net/Articles/943758/ might help#asahi-gpu · 2023-09-03 15:39
jannaupasses deqp-runner deqp_{egl,es2,es3,es31} on m1 ultra. previously segfaulting programs still work#asahi-gpu · 2023-09-03 13:03
jannaufix confirmed on asahi alarm#asahi-gpu · 2023-09-03 09:00
jannausame as reported with supertuxkart in https://github.com/AsahiLinux/linux/issues/72#issuecomment-1703891618#asahi-gpu · 2023-09-03 08:48
jannauseen with prusa-slicer#asahi-gpu · 2023-09-03 08:48
jannaufixed by reverting "bf4d2a40bada asahi: Force staging copy for LINEAR buffer reads from WC BOs"#asahi-gpu · 2023-09-03 08:45
jannauresult in a non-debug build is a segfault in disk_cache_compute_key()#asahi-gpu · 2023-09-03 08:44
jannauthe agx_linear_allowed one in agx_select_best_modifier()#asahi-gpu · 2023-09-03 08:43
jannauasahi-20230902 hits an assert while allocating a staging resource for a compressed texture (DXT5_RGBA)#asahi-gpu · 2023-09-01 11:07
jannaulina: segfault https://paste.debian.net/1290653/ but I'll try first the branch from #184#asahi-gpu · 2023-08-21 11:48
jannaueric_engestrom: thanks for https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24808 - I was going to look/ask how that should be handled